Enhancing individual glomerular filtration rate assessment: can we trust the equation? Development and validation of machine learning models to assess the trustworthiness of estimated GFR compared to measured GFR.
| Title: | Enhancing individual glomerular filtration rate assessment: can we trust the equation? Development and validation of machine learning models to assess the trustworthiness of estimated GFR compared to measured GFR. |
|---|---|
| Authors: | Lanot, Antoine, Akesson, Anna, Nakano, Felipe Kenji, Vens, Celine, Björk, Jonas, Nyman, Ulf, Grubb, Anders, Sundin, Per-Ola, Eriksen, Björn O., Melsom, Toralf, Rule, Andrew D., Berg, Ulla, Littmann, Karin, Åsling-Monemi, Kajsa, Hansson, Magnus, Larsson, Anders, Courbebaisse, Marie, Dubourg, Laurence, Couzi, Lionel, Gaillard, Francois |
| Source: | BMC Nephrology; 1/30/2025, Vol. 26 Issue 1, p1-13, 13p |
| Subject Terms: | MACHINE learning, GLOMERULAR filtration rate, RANDOM forest algorithms, TRUST, KIDNEY physiology |
| Abstract: | Background: Creatinine-based estimated glomerular filtration rate (eGFR) equations are widely used in clinical practice but exhibit inherent limitations. On the other side, measuring GFR is time consuming and not available in routine clinical practice. We developed and validated machine learning models to assess the trustworthiness (i.e. the ability of equations to estimate measured GFR (mGFR) within 10%, 20% or 30%) of the European Kidney Function Consortium (EKFC) equation at the individual level. Methods: This observational study used data from European and US cohorts, comprising 22,343 participants of all ages with available mGFR results. Four machine learning and two traditional logistic regression models were trained on a cohort of 9,202 participants to predict the likelihood of the EKFC creatinine-derived eGFR falling within 30% (p30), 20% (p20) or 10% (p10) of the mGFR value. The algorithms were internally and then externally validated on cohorts of respectively 3,034 and 10,107 participants. The predictors included in the models were creatinine, age, sex, height, weight, and EKFC. Results: The random forest model was the most robust model. In the external validation cohort, the model achieved an area under the curve of 0.675 (95%CI 0.660;0.690) and an accuracy of 0.716 (95%CI 0.707;0.725) for the P30 criterion. Sensitivity was 0.756 (95%CI 0.747;0.765) and specificity was 0.485 (95%CI 0.460; 0.511) at the 80% probability level that EKFC falls within 30% of mGFR. At the population level, the PPV of this machine learning model was 89.5%, higher than the EKFC P30 of 85.2%. A free web-application was developed to allow the physician to assess the trustworthiness of EKFC at the individual level. Conclusions: A strategy using machine learning model marginally improves the trustworthiness of GFR estimation at the population level. An additional value of this approach lies in its ability to provide assessments at the individual level. [ABSTRACT FROM AUTHOR] |
| Copyright of BMC Nephrology is the property of BioMed Central and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Complementary Index |
| FullText | Text: Availability: 0 CustomLinks: – Url: https://resolver.ebsco.com/c/xy5jbn/result?sid=EBSCO:edb&genre=article&issn=14712369&ISBN=&volume=26&issue=1&date=20250130&spage=1&pages=1-13&title=BMC Nephrology&atitle=Enhancing%20individual%20glomerular%20filtration%20rate%20assessment%3A%20can%20we%20trust%20the%20equation%3F%C2%A0Development%20and%20validation%20of%20machine%20learning%20models%20to%20assess%20the%20trustworthiness%20of%20estimated%20GFR%20compared%20to%20measured%20GFR.&aulast=Lanot%2C%20Antoine&id=DOI:10.1186/s12882-025-03972-0 Name: Full Text Finder (for New FTF UI) (s8985755) Category: fullText Text: Find It @ SCU Libraries MouseOverText: Find It @ SCU Libraries |
|---|---|
| Header | DbId: edb DbLabel: Complementary Index An: 182565485 RelevancyScore: 1082 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 1082.12060546875 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Enhancing individual glomerular filtration rate assessment: can we trust the equation? Development and validation of machine learning models to assess the trustworthiness of estimated GFR compared to measured GFR. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Lanot%2C+Antoine%22">Lanot, Antoine</searchLink><br /><searchLink fieldCode="AR" term="%22Akesson%2C+Anna%22">Akesson, Anna</searchLink><br /><searchLink fieldCode="AR" term="%22Nakano%2C+Felipe+Kenji%22">Nakano, Felipe Kenji</searchLink><br /><searchLink fieldCode="AR" term="%22Vens%2C+Celine%22">Vens, Celine</searchLink><br /><searchLink fieldCode="AR" term="%22Björk%2C+Jonas%22">Björk, Jonas</searchLink><br /><searchLink fieldCode="AR" term="%22Nyman%2C+Ulf%22">Nyman, Ulf</searchLink><br /><searchLink fieldCode="AR" term="%22Grubb%2C+Anders%22">Grubb, Anders</searchLink><br /><searchLink fieldCode="AR" term="%22Sundin%2C+Per-Ola%22">Sundin, Per-Ola</searchLink><br /><searchLink fieldCode="AR" term="%22Eriksen%2C+Björn+O%2E%22">Eriksen, Björn O.</searchLink><br /><searchLink fieldCode="AR" term="%22Melsom%2C+Toralf%22">Melsom, Toralf</searchLink><br /><searchLink fieldCode="AR" term="%22Rule%2C+Andrew+D%2E%22">Rule, Andrew D.</searchLink><br /><searchLink fieldCode="AR" term="%22Berg%2C+Ulla%22">Berg, Ulla</searchLink><br /><searchLink fieldCode="AR" term="%22Littmann%2C+Karin%22">Littmann, Karin</searchLink><br /><searchLink fieldCode="AR" term="%22Åsling-Monemi%2C+Kajsa%22">Åsling-Monemi, Kajsa</searchLink><br /><searchLink fieldCode="AR" term="%22Hansson%2C+Magnus%22">Hansson, Magnus</searchLink><br /><searchLink fieldCode="AR" term="%22Larsson%2C+Anders%22">Larsson, Anders</searchLink><br /><searchLink fieldCode="AR" term="%22Courbebaisse%2C+Marie%22">Courbebaisse, Marie</searchLink><br /><searchLink fieldCode="AR" term="%22Dubourg%2C+Laurence%22">Dubourg, Laurence</searchLink><br /><searchLink fieldCode="AR" term="%22Couzi%2C+Lionel%22">Couzi, Lionel</searchLink><br /><searchLink fieldCode="AR" term="%22Gaillard%2C+Francois%22">Gaillard, Francois</searchLink> – Name: TitleSource Label: Source Group: Src Data: BMC Nephrology; 1/30/2025, Vol. 26 Issue 1, p1-13, 13p – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22MACHINE+learning%22">MACHINE learning</searchLink><br /><searchLink fieldCode="DE" term="%22GLOMERULAR+filtration+rate%22">GLOMERULAR filtration rate</searchLink><br /><searchLink fieldCode="DE" term="%22RANDOM+forest+algorithms%22">RANDOM forest algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22TRUST%22">TRUST</searchLink><br /><searchLink fieldCode="DE" term="%22KIDNEY+physiology%22">KIDNEY physiology</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Background: Creatinine-based estimated glomerular filtration rate (eGFR) equations are widely used in clinical practice but exhibit inherent limitations. On the other side, measuring GFR is time consuming and not available in routine clinical practice. We developed and validated machine learning models to assess the trustworthiness (i.e. the ability of equations to estimate measured GFR (mGFR) within 10%, 20% or 30%) of the European Kidney Function Consortium (EKFC) equation at the individual level. Methods: This observational study used data from European and US cohorts, comprising 22,343 participants of all ages with available mGFR results. Four machine learning and two traditional logistic regression models were trained on a cohort of 9,202 participants to predict the likelihood of the EKFC creatinine-derived eGFR falling within 30% (p30), 20% (p20) or 10% (p10) of the mGFR value. The algorithms were internally and then externally validated on cohorts of respectively 3,034 and 10,107 participants. The predictors included in the models were creatinine, age, sex, height, weight, and EKFC. Results: The random forest model was the most robust model. In the external validation cohort, the model achieved an area under the curve of 0.675 (95%CI 0.660;0.690) and an accuracy of 0.716 (95%CI 0.707;0.725) for the P30 criterion. Sensitivity was 0.756 (95%CI 0.747;0.765) and specificity was 0.485 (95%CI 0.460; 0.511) at the 80% probability level that EKFC falls within 30% of mGFR. At the population level, the PPV of this machine learning model was 89.5%, higher than the EKFC P30 of 85.2%. A free web-application was developed to allow the physician to assess the trustworthiness of EKFC at the individual level. Conclusions: A strategy using machine learning model marginally improves the trustworthiness of GFR estimation at the population level. An additional value of this approach lies in its ability to provide assessments at the individual level. [ABSTRACT FROM AUTHOR] – Name: Abstract Label: Group: Ab Data: <i>Copyright of BMC Nephrology is the property of BioMed Central and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://login.libproxy.scu.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edb&AN=182565485 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1186/s12882-025-03972-0 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 13 StartPage: 1 Subjects: – SubjectFull: MACHINE learning Type: general – SubjectFull: GLOMERULAR filtration rate Type: general – SubjectFull: RANDOM forest algorithms Type: general – SubjectFull: TRUST Type: general – SubjectFull: KIDNEY physiology Type: general Titles: – TitleFull: Enhancing individual glomerular filtration rate assessment: can we trust the equation? Development and validation of machine learning models to assess the trustworthiness of estimated GFR compared to measured GFR. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Lanot, Antoine – PersonEntity: Name: NameFull: Akesson, Anna – PersonEntity: Name: NameFull: Nakano, Felipe Kenji – PersonEntity: Name: NameFull: Vens, Celine – PersonEntity: Name: NameFull: Björk, Jonas – PersonEntity: Name: NameFull: Nyman, Ulf – PersonEntity: Name: NameFull: Grubb, Anders – PersonEntity: Name: NameFull: Sundin, Per-Ola – PersonEntity: Name: NameFull: Eriksen, Björn O. – PersonEntity: Name: NameFull: Melsom, Toralf – PersonEntity: Name: NameFull: Rule, Andrew D. – PersonEntity: Name: NameFull: Berg, Ulla – PersonEntity: Name: NameFull: Littmann, Karin – PersonEntity: Name: NameFull: Åsling-Monemi, Kajsa – PersonEntity: Name: NameFull: Hansson, Magnus – PersonEntity: Name: NameFull: Larsson, Anders – PersonEntity: Name: NameFull: Courbebaisse, Marie – PersonEntity: Name: NameFull: Dubourg, Laurence – PersonEntity: Name: NameFull: Couzi, Lionel – PersonEntity: Name: NameFull: Gaillard, Francois IsPartOfRelationships: – BibEntity: Dates: – D: 30 M: 01 Text: 1/30/2025 Type: published Y: 2025 Identifiers: – Type: issn-print Value: 14712369 Numbering: – Type: volume Value: 26 – Type: issue Value: 1 Titles: – TitleFull: BMC Nephrology Type: main |
| ResultId | 1 |